A crusher type mentioned above is utilized in order to crush hard material, such as pieces of rock material.
It is desirable to be able to crush a large quantity of material in the crusher without risking the crusher being exposed to loads which cause breakdowns. High productivity leads to high crushing forces or high crushing power. Especially, when considering the crushing force and the crushing power, the crusher may be damaged due to frequent peak levels of force and power experienced by the crusher.
Document WO-2005/007293-A1 discloses a method for controlling a crusher. In the method, the instantaneous load on the crusher is measured during a period of time, a reference value representing the highest measured instantaneous load during the period of time is calculated, and the representative value is compared to a predetermined value. On the basis of the results of the comparison, the loading on the crusher is controlled. However, it is difficult to determine the reference value, and therefore unnecessary delays may occur. Also, the representative value does not necessarily show the actual number of the crusher load peaks affecting the control system and structure of the crusher.
Document WO 87/05828 discloses a method to decrease the risk of increased mechanical load and breakdowns. The pressure surges in the hydraulic system are monitored, and when a predetermined value is exceeded, the relative position of the crushing shells is changed and the width of the crushing gap is increased. The disclosed method reduces the risk of the crusher breaking down, but can not be used to improve the performance of the crusher. Additionally, the method disclosed in document WO 87/05828 does not explain how to control the crusher to restore the normal operation after the gap width has been increased.
Furthermore, excessive calculations and the averaging the measured values of power and force to reduce random variation, may result in a delayed response to overload conditions. For example, delays occur due to the way in which averages are calculated by combining the current reading with either a past average value or with a fixed set of past readings. The averaging may also wash out the range of values and so peak levels are ignored and this can cause a problem.